Method for pre-judging collapse of blast furnace based on static pressure

Abstract

The application discloses a method for pre-judging blast furnace material collapse based on static pressure, which comprises the following steps: collecting data of a static pressure measuring point by a computer; linearly fitting a real-time static pressure measuring point value and a static pressure measuring point elevation to obtain a function, a slope of the function is used as a pressure gradient reflecting a blast furnace gas flow and a material feeding feature to draw a pressure gradient change curve with time, and when a curve at a current time t satisfies condition A and condition B, it is determined that the blast furnace has a possibility of material collapse. The application adopts a linear regression method for the static pressure to obtain the pressure gradient, compared with directly analyzing the static pressure or a segmented pressure difference, the analysis of the pressure gradient greatly reduces the difficulty and workload of a technical personnel in analyzing the blast furnace gas flow, in addition, the condition A and the condition B for pre-judging the material collapse are combined by using the "and" mode, the long-period analysis and the instantaneous value analysis of the pressure gradient are taken into account, and the accuracy and the scientificity of the pre-judgment of the material collapse are ensured.

Description

Technical Field

[0001] This invention relates to the field of metallurgical technology, specifically to a method for predicting blast furnace collapse based on static pressure. Background Technology

[0002] In blast furnace ironmaking, a uniform, orderly, uninterrupted, and non-collapsing descent of the burden surface is a crucial indicator of normal furnace conditions. A rapid descent of the burden surface exceeding 1.0m is termed a collapse. Collapse causes a rapid drop in blast furnace temperature and disrupts the gas flow distribution. If collapse is continuous, the furnace conditions are highly likely to deteriorate into serious abnormalities, even leading to hearth freezing. Blast furnace technicians must strive to minimize the occurrence or severity of collapse while controlling furnace conditions. Technicians typically predict collapse based on blast volume, hot blast pressure, probe data, and burden velocity, and take proactive measures. However, such predictions are often subjective and unstable; sometimes, atypical furnace condition characteristics preceding collapse are overlooked, leading to prediction failures.

[0003] Chinese Patent No. CN202311059721.5 discloses an intelligent prediction and response method for blast furnace collapse, including the following steps: (1) calculating the average daily oxygen consumption per ton of iron charged in the blast furnace; (2) correcting the current oxygen consumption per ton of iron charged according to the current fuel ratio; (3) calculating the difference Xi between the theoretical and actual iron charge at the current working moment of the blast furnace, and judging the internal charging space of the blast furnace in real time; (4) calculating the collapse index ηi in real time and assessing the collapse risk; (5) evaluating the collapse index. This method can accurately predict blast furnace collapse, prevent collapses in advance, and reduce collapse accidents. A blast furnace collapse prediction method disclosed in patent number CN201910811704.X includes: acquiring a reference image; acquiring a target image, performing image matching with the reference image, and confirming whether the acquired target image indicates a collapse trend; wherein, multiple features of the image before collapse are extracted for model training, and the training samples include the target image, the reference image, the image contour clarity, the brightness information, and the matching degree. The target image and the reference image are used as training inputs, and the image contour, brightness information, and matching degree are used as output reference values. The training samples are trained using a generative adversarial network (GAN), training the initial first convolutional layer, the initial second convolutional layer, and the initial GAN ​​to obtain the trained first convolutional layer, the second convolutional layer, and the GAN. This method can quickly and accurately analyze whether the current stage of blast furnace operation has a collapse trend. Although the above patent can predict collapse trends, its calculation process is complex, the program response time is long, and the hardware requirements are high, while also increasing the difficulty and workload for technicians in analyzing the gas flow inside the furnace. Summary of the Invention

[0004] The purpose of this invention is to provide a method for predicting blast furnace collapse based on static pressure. By using linear regression of static pressure to obtain the pressure gradient, compared with directly analyzing static pressure or segmented pressure difference, analyzing the pressure gradient greatly reduces the difficulty and workload of technicians in analyzing the gas flow inside the furnace, thereby solving the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for predicting blast furnace collapse based on static pressure includes the following steps:

[0007] S1: Collect a set of static pressure measurement points every 10-60 seconds using a computer;

[0008] S2: Simultaneously, the real-time static pressure measurement points and their elevations are linearly fitted to obtain the normal function. The slope of the normal function is used as the pressure gradient based on the airflow and feed characteristics within the reactor. Unit: kPa / m;

[0009] S3: Draw the pressure gradient If the curve changes over time, and at the current time t, the curve satisfies both condition A and condition B, it is determined that there is a possibility of a blast furnace collapse.

[0010] Furthermore, the values ​​and elevations of the static pressure measuring points in S2 are obtained by linearly fitting the real-time data of the static pressure measuring points within the height range from the lower boundary of the furnace waist to the 2 / 3 height line of the furnace body.

[0011] Furthermore, in S2, the independent variable of the normal function is the elevation of the static pressure measuring point, and the dependent variable is the value of the static pressure measuring point.

[0012] Furthermore, in S3, condition A represents the pressure gradient over the time period t0 to t20 min. The mean absolute value of the pressure gradient over the time period t20min to t60min The mean absolute value is more than 0.5 kPa / m less.

[0013] Furthermore, in S3, condition B is the pressure gradient during the time period t0 to t20 min. The absolute value of is decreasing, and the minimum value appears later than the maximum value. The minimum value is more than 1.0 kPa / m smaller than the maximum value.

[0014] Compared with the prior art, the beneficial effects of the present invention are:

[0015] The present invention provides a method for predicting blast furnace collapse based on static pressure. This method uses linear regression of static pressure to obtain the pressure gradient. Compared with directly analyzing static pressure or segmented pressure difference, analyzing the pressure gradient greatly reduces the difficulty and workload for technicians in analyzing the gas flow inside the furnace. In addition, the conditions A and B for predicting collapse are combined using an AND statement, taking into account both long-term and instantaneous value analysis of the pressure gradient, thus ensuring the accuracy and scientific nature of the predicted collapse. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of linear fitting of static pressure data and static pressure measuring point elevation in an embodiment of the present invention;

[0017] Figure 2 This is a pressure gradient curve of the blast furnace during the 266th minute to the 348th minute of operation on a certain day, as shown in this embodiment of the invention.

[0018] Figure 3 This is a probe curve of the blast furnace during the 266th minute to the 348th minute of operation on a certain day, as shown in this embodiment of the invention.

[0019] Figure 4 This is a graph showing the air volume curve of the blast furnace during the 266th to 348th minute of operation on a certain day, as described in an embodiment of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] This invention provides a method for predicting blast furnace collapse based on static pressure, comprising the following steps:

[0022] Step 1: Collect a set of static pressure measurement points every 10-60 seconds using a computer;

[0023] Step 2: Simultaneously perform linear fitting between the real-time static pressure measurement points and their elevations to obtain the normal function. The slope of the normal function is used as the pressure gradient based on the airflow and feed characteristics within the reactor. The unit is kPa / m; the static pressure measurement point value and elevation are selected from the real-time data of the static pressure measurement points within the height range from the lower boundary of the furnace waist to the 2 / 3 height line of the furnace body, and linearly fitted; the independent variable of the normal function is the elevation of the static pressure measurement point, and the dependent variable is the value of the static pressure measurement point;

[0024] Step 3: Draw the pressure gradient The curves changing over time indicate that if the curve at the current time t satisfies both condition A and condition B, it is determined that the blast furnace is likely to experience a collapse. Condition A is the pressure gradient over the time period t0 to t20 minutes. The mean absolute value of the pressure gradient over the time period t20min to t60min The mean absolute value is less than 0.5 kPa / m or more; condition B is the pressure gradient within the time period t0 to t20 min. The absolute value of is decreasing, and the minimum value appears later than the maximum value. The minimum value is more than 1.0 kPa / m smaller than the maximum value.

[0025] To further illustrate the embodiments of the present invention, the following specific examples are provided:

[0026] Please see Figure 1-4 With a certain 2500m 3 The static pressure measuring points for the blast furnace are configured as shown in Table 1 below. The computer collects a set of static pressure data every 10 seconds.

[0027] Table 1 Static pressure measurement points

[0028] Pressure measurement points Elevation (m) Number of measuring points 5th layer static pressure 30.665 6 4th layer static pressure 27.010 6 Static pressure of the third layer 23.552 6 Static pressure of the second layer 20.050 6 Static pressure of the first layer 16.932 6

[0029] The lower boundary of the furnace waist is 19.000m, and the elevation of the 2 / 3 height line of the furnace body is 32.533m. Therefore, the static pressure data of the first layer of cooling wall is not included in the linear fitting calculation.

[0030] The real-time static pressure measurement values ​​from layers 2 to 5 are linearly fitted with the static pressure measurement point elevations. The independent variable is the elevation of the static pressure measurement point, and the dependent variable is the static pressure measurement value. The slope of the normal function is the gradient of the fitted static pressure along the blast furnace height direction. The unit is kPa / m, as shown in Table 2 below.

[0031] Table 2 Real-time static pressure data and fitted gradient

[0032]

[0033] Gradients are made within the secondary system of the blast furnace. The curve changes over time. When the curve reaches the 308th minute: (1) Pressure gradient during the time period 288-308 minutes. The absolute mean of the pressure gradient over the time period of 248 min–288 min The mean absolute value is less than 0.50036 kPa / m; (2) Pressure gradient during the time period of 288 min-308 min. The absolute value of is decreasing, and the minimum value appears later than the maximum value. The minimum value is 1.22 kPa / m smaller than the maximum value.

[0034] This also triggered condition A (the pressure gradient within the time period t0 to t20 min). The mean absolute value of the pressure gradient over the time period t20min to t60min The absolute value of the mean is less than 0.5 kPa / m and condition B (pressure gradient within the time period t0 to t20 min). If the absolute value of the blast furnace is decreasing and the minimum value appears later than the maximum value (the minimum value is more than 1.0 kPa / m smaller than the maximum value), the system determines that the blast furnace may collapse.

[0035] The blast furnace foreman reduced the blast by 100m at 311 minutes. 3 During the 326th minute of operation, a blast furnace collapse occurred, reaching a depth of 1.3 meters. The foreman reduced the airflow 15 minutes prior to the collapse, mitigating its severity and consequences.

[0036] In summary, this invention provides a method for predicting blast furnace collapse based on static pressure. To predict collapse, technicians have reduced the workload and difficulty by analyzing only one pressure gradient, instead of more than 20 key parameters. Furthermore, after implementing this invention, the secondary system can provide early warning of collapse 15-30 minutes in advance. Technicians can take preventative measures to avoid or mitigate collapse, thus helping to maintain smooth furnace operation. In addition, because this invention fully explores the significance of static pressure detection in segmented monitoring of pressure differences within the blast furnace, technicians have generally reported a deeper understanding of segmented and localized airflow changes within the furnace after using this method.

[0037] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for predicting blast furnace collapse based on static pressure, characterized in that, Includes the following steps: S1: Collect a set of static pressure measurement points every 10-60 seconds using a computer; S2: Simultaneously, the real-time static pressure measurement point values ​​and static pressure measurement point elevations are linearly fitted to obtain the normal function. The slope of the normal function is used as the pressure gradient ∇ of the airflow and material feeding characteristics in the reactor, in kPa / m. S3: Plot the curve of pressure gradient ∇ over time. If the curve at the current time t satisfies both condition A and condition B, it is determined that there is a possibility of blast furnace collapse. In S3, condition A is that the mean absolute value of the pressure gradient ∇ during the time period t0 to t20 min is more than 0.5 kPa / m smaller than the mean absolute value of the pressure gradient ∇ during the time period t20 min to t60 min. In S3, condition B is that the absolute value of the pressure gradient ∇ during the time period t0 to t20 min is decreasing, and the minimum value appears later than the maximum value, and the minimum value is more than 1.0 kPa / m smaller than the maximum value.

2. The method for predicting blast furnace collapse based on static pressure as described in claim 1, characterized in that: The values ​​and elevations of the static pressure measuring points in S2 are obtained by linearly fitting the real-time data of the static pressure measuring points within the height range from the lower boundary of the furnace waist to the 2 / 3 height line of the furnace body.

3. The method for predicting blast furnace collapse based on static pressure as described in claim 1, characterized in that: In S2, the independent variable of the normal function is the elevation of the static pressure measuring point, and the dependent variable is the value of the static pressure measuring point.

Images